Routine molecular screening of advanced refractory cancer patients: feasible, but beneficial only for some

Genomic testing of tumor samples can enable personalized treatment selection, when targeted treatments are matched to genetic changes in the tumor. Although a growing number of patients with advanced cancer receives some genomic testing, comprehensive genomic testing is not yet routine care. A French study (ProfiLER) including 1,944 patients with advanced cancer suggests that widespread, routine genomic testing is feasible, although it offers clinical benefit to only a subset of patients at this time. The researchers found at least one “actionable” alteration in 52% of the tumor samples tested (“actionable” alterations are genetic changes that theoretically can be matched to targeted therapies).

A molecular tumor experts board recommended targeted treatments to 676 patients, and 143 of them actually received the recommended treatment. Interestingly, the five-year survival rate was higher among patients who received the recommended treatment than among those who did not (34.8% vs. 28.1%).

ProfiLER is an ongoing clinical trial that uses genomic profiling of tumors to guide treatment decisions for patients with advanced cancer. DNA from tumor samples is analyzed by next-generation sequencing of 69 cancer-related genes and by whole-genome-comparative genomic hybridization, two widely available technologies. A multidisciplinary board of experts (the ‘molecular tumor board’) meets weekly to review genomic testing results and, if actionable mutations are found, provides recommendations for molecularly targeted therapies. To date, 2,676 patients have enrolled in the study, and 1,944 tumors were analyzed, including colorectal-, gynaecologic-, breast-, brain-, and head and neck cancer, as well as sarcoma. Actionable mutations were found in 1,004 (52%) tumor samples. 609 patients of whom the samples were derived, had only one actionable mutation, and 394 had two or more (up to six). The most common actionable mutations were found in KRAS (N= 156; 8.5%), PIK3CA (N= 150; 8.2%), CDKN2A HD (N= 174; 9.5%), PTEN HD (N= 49, 2.7%), CCND1 (N= 97; 5.3%), FGFR1 (N= 56; 3.1%), MDM2 (N= 53; 2.9%), HER2 (N= 42; 2.3%) and HER1(N= 41; 2.2%). The molecular tumor board recommended molecularly targeted treatments to 676 patients (35% of 1,944 tested) based on availability of drugs hitting either the target protein or the pathway activated by the target. Of those, 143 received the recommended treatment, most through enrollment in a clinical trial. The other 533 patients were not able to receive the recommended treatment because of poor health, rapid progression of the cancer, not meeting eligibility criteria for a clinical trial, or difficulties to obtain off-label commercial drugs. Researchers compared the overall survival rates for the 143 patients who received targeted therapies based on genomic testing and 502 patients who did not. At three years, 53.7% of patients who received the recommended targeted therapy were alive, as compared to 46.1% of patients who did not. The five-year survival rate was also higher for patients who received targeted therapy (34.8% vs. 28.1%).

This study confirms that comprehensive genomic profiling can be performed in routine practice to select patients for targeted cancer therapies. Actionable alterations were found in about half of the patients, with treatment recommendation in approximately 35%. Most patients treated derived benefit from the recommended targeted treatment, but these represent a minority of the whole population screened. The researchers are planning a new, randomized clinical trial, ProfiLER 02, which will compare the 70-gene test used in the current study to a commercial, 315-gene test. That trial should reveal whether screening a larger number of genes leads to more recommendations for targeted therapy.

Reference

Tredan O, Corset V, Wang Q, et al. Routine molecular screening of advanced refractory cancer patients: An analysis of the first 2490 patients of the ProfilER Study. Presented at ASCO 2017; Abstract LBA100.